Analytical response time estimation in parallel relational database systems

N. Tomov, Euan W. Dempster, M. H. Williams, A. Burger, H. Taylor, P. J. B. King, P. Broughton

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

Techniques for performance estimation in parallel database systems are well established for parameters such as throughput, bottlenecks and resource utilisation. However, response time estimation is a complex activity which is difficult to predict and has attracted research for a number of years. Simulation is one option for predicting response time but this is a costly process. Analytical modelling is a less expensive option but requires approximations and assumptions about the queueing networks built up in real parallel database machines which are often questionable and few of the papers on analytical approaches are backed by results from validation against real machines. This paper describes a new analytical approach for response time estimation that is based on a detailed study of different approaches and assumptions. The approach has been validated against two commercial parallel DBMSs running on actual parallel machines and is shown to produce acceptable accuracy.
Original languageEnglish
Pages (from-to)249-283
Number of pages35
JournalParallel Computing
Volume30
Issue number2
DOIs
Publication statusPublished - Feb 2004

Fingerprint

Parallel Databases
Relational database systems
Database Systems
Relational Database
Response Time
Queueing networks
Analytical Modeling
Queueing Networks
Parallel Machines
Parallel Systems
Throughput
Predict
Resources
Approximation
Simulation

Cite this

Tomov, N., Dempster, E. W., Williams, M. H., Burger, A., Taylor, H., King, P. J. B., & Broughton, P. (2004). Analytical response time estimation in parallel relational database systems. Parallel Computing, 30(2), 249-283. https://doi.org/10.1016/j.parco.2003.11.003
Tomov, N. ; Dempster, Euan W. ; Williams, M. H. ; Burger, A. ; Taylor, H. ; King, P. J. B. ; Broughton, P. / Analytical response time estimation in parallel relational database systems. In: Parallel Computing. 2004 ; Vol. 30, No. 2. pp. 249-283.
@article{e6ef901259ba4eaeb5c136921956ec16,
title = "Analytical response time estimation in parallel relational database systems",
abstract = "Techniques for performance estimation in parallel database systems are well established for parameters such as throughput, bottlenecks and resource utilisation. However, response time estimation is a complex activity which is difficult to predict and has attracted research for a number of years. Simulation is one option for predicting response time but this is a costly process. Analytical modelling is a less expensive option but requires approximations and assumptions about the queueing networks built up in real parallel database machines which are often questionable and few of the papers on analytical approaches are backed by results from validation against real machines. This paper describes a new analytical approach for response time estimation that is based on a detailed study of different approaches and assumptions. The approach has been validated against two commercial parallel DBMSs running on actual parallel machines and is shown to produce acceptable accuracy.",
author = "N. Tomov and Dempster, {Euan W.} and Williams, {M. H.} and A. Burger and H. Taylor and King, {P. J. B.} and P. Broughton",
year = "2004",
month = "2",
doi = "10.1016/j.parco.2003.11.003",
language = "English",
volume = "30",
pages = "249--283",
journal = "Parallel Computing",
issn = "0167-8191",
publisher = "Elsevier",
number = "2",

}

Tomov, N, Dempster, EW, Williams, MH, Burger, A, Taylor, H, King, PJB & Broughton, P 2004, 'Analytical response time estimation in parallel relational database systems', Parallel Computing, vol. 30, no. 2, pp. 249-283. https://doi.org/10.1016/j.parco.2003.11.003

Analytical response time estimation in parallel relational database systems. / Tomov, N.; Dempster, Euan W.; Williams, M. H.; Burger, A.; Taylor, H.; King, P. J. B.; Broughton, P.

In: Parallel Computing, Vol. 30, No. 2, 02.2004, p. 249-283.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Analytical response time estimation in parallel relational database systems

AU - Tomov, N.

AU - Dempster, Euan W.

AU - Williams, M. H.

AU - Burger, A.

AU - Taylor, H.

AU - King, P. J. B.

AU - Broughton, P.

PY - 2004/2

Y1 - 2004/2

N2 - Techniques for performance estimation in parallel database systems are well established for parameters such as throughput, bottlenecks and resource utilisation. However, response time estimation is a complex activity which is difficult to predict and has attracted research for a number of years. Simulation is one option for predicting response time but this is a costly process. Analytical modelling is a less expensive option but requires approximations and assumptions about the queueing networks built up in real parallel database machines which are often questionable and few of the papers on analytical approaches are backed by results from validation against real machines. This paper describes a new analytical approach for response time estimation that is based on a detailed study of different approaches and assumptions. The approach has been validated against two commercial parallel DBMSs running on actual parallel machines and is shown to produce acceptable accuracy.

AB - Techniques for performance estimation in parallel database systems are well established for parameters such as throughput, bottlenecks and resource utilisation. However, response time estimation is a complex activity which is difficult to predict and has attracted research for a number of years. Simulation is one option for predicting response time but this is a costly process. Analytical modelling is a less expensive option but requires approximations and assumptions about the queueing networks built up in real parallel database machines which are often questionable and few of the papers on analytical approaches are backed by results from validation against real machines. This paper describes a new analytical approach for response time estimation that is based on a detailed study of different approaches and assumptions. The approach has been validated against two commercial parallel DBMSs running on actual parallel machines and is shown to produce acceptable accuracy.

U2 - 10.1016/j.parco.2003.11.003

DO - 10.1016/j.parco.2003.11.003

M3 - Article

VL - 30

SP - 249

EP - 283

JO - Parallel Computing

JF - Parallel Computing

SN - 0167-8191

IS - 2

ER -